Executive Interviews

Executive Interview: Simon Hunt

In an exclusive interview with The Executive Magazine, Simon Hunt, Co-founder and CEO of Brego, reveals how his Goldman Sachs background and personal experience with car depreciation led to creating revolutionary AI-driven vehicle valuation technology that now serves prestigious lenders and dealerships across the UK

Elizabeth Jenkins-Smalley

Editor In Chief at The Executive Magazine

Share this article:

Simon Hunt’s journey from Goldman Sachs lead developer to revolutionary entrepreneur reads like a masterclass in identifying market inefficiencies and leveraging cutting-edge technology to solve them. As Co-founder and CEO of Brego, Hunt has transformed the traditionally conservative vehicle valuation industry by deploying sophisticated neural networks that can predict values for everything from Toyota Aygos to McLaren P1s—even for vehicles that haven’t been released yet. In this exclusive interview with The Executive Magazine, Hunt shares the pivotal moments that led him to abandon a lucrative career in financial services to pursue his vision of bringing artificial intelligence to automotive valuations.

What began as a personal frustration with car depreciation in 2016 has evolved into a technology platform trusted by prestigious institutions including JBR Capital, United Trust Bank, and high-end dealerships such as Romans International. Hunt’s unique combination of financial modelling expertise and deep understanding of automotive markets has enabled Brego to achieve what traditional valuation providers have struggled with for decades: delivering accuracy and coverage across the entire spectrum of vehicles. As the automotive industry faces unprecedented disruption from electric vehicle adoption and changing consumer preferences, Hunt’s AI-driven approach positions him at the forefront of a transformation that could reshape how the industry values and trades vehicles for years to come.

Having spent years as a lead developer at Goldman Sachs working with sophisticated financial modelling systems, what drove your decision to leave an established career in financial services to pursue entrepreneurship in the automotive sector?

“I’ve always wanted to build a company, but I wanted to solve a problem I was passionate about that could make real use of new technology. We saw incumbent valuation providers using manual methods and generic curves to match large samples of vehicles and where newer providers used technology, the techniques often struggled and didn’t allow valuations for unreleased models or more specialist vehicles.”

Traditional vehicle valuation providers like CAP HPI and Glass’s Guide have dominated the UK market for over 30 years using established data analysis methods. What fundamental differences exist between your AI approach and these conventional systems, and why do you believe artificial intelligence can deliver more accurate vehicle valuations than traditional methods?

“There are experts in the industry who can value classic Ferraris by name, studying their condition and understanding that market deeply. Some mass-market cars do behave in repeatable patterns when it comes to depreciation and simple curves can occasionally work. Unfortunately, that’s not the case for the majority of vehicles. Certain features of cars have impact due to emotional, social or economic factors and no human can retain millions of historic sales data points (across trims, options, mileage, regions and seasons) well enough to be an expert on every make and model; AI can.

“We’ve seen the same with systems like ChatGPT, Gemini and others. AI can internalise far more information than any human, which makes it an excellent tool for analysing huge datasets and ‘learning’ what the trends are and why they happen. This is why, in our experience and client back-tests, AI delivers more accurate valuations than traditional curve-based approaches; there’s simply too much data and too many interactions for other methods. Continuous retraining and back-testing also help us adapt faster to market shifts, and we provide confidence ranges and explainability so teams can trust the output.”

Building a technology company alongside your brother Philip presents unique dynamics in leadership and decision-making. How have you structured the co-founder relationship to leverage your respective expertise in product management and development, and what governance frameworks have proven most effective in maintaining both family harmony and business rigour?

“I firmly believe that my brother, Philip, is one of the best data engineers in the UK. His ability to understand big-data problems and develop solutions is second to none. He learnt how our neural networks function and has led the technical team since we started the company together. My skill set is more in product management and design and, by chance, we specialise in opposite areas of running a business, so we rarely step on each other’s toes. That’s made building Brego hard work but straightforward in terms of leveraging our respective expertise and splitting responsibilities to develop the product and the business.

“We keep decision rights clear: Philip owns data/ML and platform engineering; I own product, design and the commercial roadmap but every decision is driven by consensus. We’ve always been close and make sure work isn’t the only thing we talk about. Even though it occupies most of our time, we keep one rule: we’re brothers first, colleagues second.”

Your initial data science project evolved into serving major financial institutions when JBR Capital articulated a crucial need for comprehensive vehicle valuations spanning from Toyota Aygos to McLaren P1s. What were the pivotal moments that transformed this market opportunity into a scalable business operation?

“Then, two pivotal developments really launched the product and business. In 2019, JBR Capital, one of our initial clients, articulated a crucial need: a web-based system capable of predicting vehicle values across the entire spectrum, from a Toyota Aygo to a McLaren P1. This highlighted the clear market demand for such a product. We proved feasibility after hiring AI engineers with a government grant and found that our AI models valued every car in our dataset with high accuracy. As more lenders and brokers came on board, leaving my full-time job felt like the obvious next step and an opportunity I didn’t want to miss.”

Your client portfolio includes prestigious lenders such as JBR Capital, United Trust Bank, and dealerships like Romans International. What specific validation challenges did you encounter when convincing established financial institutions to adopt AI-driven valuations over traditional methods, and how did you demonstrate the superior accuracy of your predictive models?

“Each customer we take on rightly tests our accuracy claims, and I believe we’ve won the vast majority of those head-to-head evaluations. The majority of lenders, dealers and other automotive businesses are actually open to new technology and we’ve been pleasantly surprised at the willingness shown across the industry to try new approaches. Generally, accuracy tests involve us valuing large volumes of vehicles, and customers then analyse the results; comparing our predictions to actual sale prices and to other valuation providers. These tests span hundreds to, in some cases, millions of vehicles, across asset types and price points and we consistently deliver lower error and tighter confidence ranges than the alternatives.

“Our platform first gained traction with lenders, which then attracted finance brokers such as Apollo Capital, Magnitude and Charles & Dean. That momentum drew in dealers seeking similar data. Today, we proudly serve a diverse clientele from supercar dealerships and lenders to major holiday-home operators – and we’re trusted for market-leading accuracy.”

The automotive industry is experiencing significant disruption with electric vehicle adoption, autonomous driving technology, and changing consumer preferences. How does your AI system account for these unprecedented market variables when generating future value predictions, particularly for vehicle categories that lack extensive historical depreciation data?

“The industry landscape is changing dramatically. Governments are setting ambitious timelines and incentives, and with the ZEV mandate in full force, valuations for some segments could come under pressure. In my lifetime, we’ve seen one example where a fuel type (diesel) was heavily promoted and later reassessed. Since then, diesel residuals have generally declined year-on-year, and some EV segments may see similar pressure, though for different reasons. As more EVs are pre-registered to help manufacturers meet targets and reduce the risk of fines, more vehicles become nearly-new stock, and the resulting supply can outpace demand in places; consumer adoption also varies by charging access, total cost of ownership and incentives.

“These market-wide shifts are incredibly difficult to predict and many millions of pounds are spent by businesses to do just that; at Brego we focus our scope on the automotive market specifically. While it’s a tough problem to solve, we are actively working with key businesses in the industry to model what 2030 and beyond will look like. The effects on the market are already being picked up by our neural network, and you can visually see changing patterns across petrol/diesel and EV segments as we move toward 2030.”

Seven years into the venture, you describe building a tech company first that happens to work in automotive. What qualities do you prioritise when assembling your team, and how do you identify individuals who can thrive in solving complex AI challenges regardless of their specific technical background?

“We always wanted Brego to be a tech company first that just so happens to work within the automotive industry. Our team is made up of the types of people we loved working with at the tech companies we’ve worked at in the past. Both Philip and I love working with people who enjoy solving problems, be it technical, product, or business-related, and not everyone on our team is from a technical background. We strive to hire people who love learning and facing complex challenges; people with that mindset can learn new technology, so it matters less which specific tools they’ve used and more that they can learn quickly.”

With growing client demand and expanding market opportunities, how have you structured your development operations to simultaneously accelerate innovation in AI capabilities whilst ensuring seamless service delivery to your established client base?

“The whole Brego team meets every week to discuss the latest product decisions and ideas. We love getting every member involved, as the most interesting ideas often come from discussion and time spent together – even though this is challenging as a remote-based company. We proactively choose what to work on as a team, and each developer has a say in what they would like to do.

“We all believe that design and user experience are at the heart of everything we do and every feature on our Platform needs to be as intuitive as possible. If a user gets confused or lost in a feature, we’ve failed in our mission to deliver it properly.

“Our APIs handle tens of millions of calls per month, and we design for high availability. Here’s where being a tech company first helps. We use a zero-downtime deployment approach, which means that even when we push major updates, customers experience no planned outage.”

Your origin story references the emotional connection between car enthusiasts and their vehicles, drawing parallels to the bond between Aragorn and Brego in Tolkien’s work. How does this philosophy of understanding the emotional aspect of vehicle ownership influence your technical approach to creating purely data-driven valuation models?

“I think it helps to understand what some cars mean to people. We’re losing sight of what cars used to mean to us. When I sold the Jaguar I was lucky enough to own, which set Brego’s journey in motion, I remember feeling like I’d lost a best friend. This emotional connection between us and cars may be fading in some areas, but for certain cars it remains strong, and the data we use to train our AI models reflects this; our AI learns which vehicles tend to command stronger demand and can hold their value better. We capture that through demand signals such as scarcity, option desirability and community interest; still purely data-driven, just informed by how people feel.

“In the early days of Brego, that understanding made testing certain cars easier, and for me personally, the love of cars makes each day a joy. We get to work with the most exciting technology while looking at vehicles that will form an important part of someone’s life, emotional or simply practical.

“I do hope that we as a society continue to have an emotional connection to vehicles. For me, the places they unlock, the journeys they take us on, and the memories they create matter. It doesn’t really matter what you drive as well, I have as many joyous memories as a student, driving a Hyundai i10 as I do now with the cars I’m lucky enough to drive and as we transition into the new world of automotive, I sincerely hope that the human, emotional side is never lost.”

You’ve highlighted accuracy and coverage as your core competitive advantages, including the ability to value unreleased vehicles like the Jaguar Type-00. What specific technological capabilities enable this breadth of valuation, and how do you anticipate the competitive landscape evolving as more companies adopt AI?

“Accuracy and Coverage. They are the two core unique selling points of Brego over other valuation providers. The use of AI allows us to predict the valuation of vehicles that aren’t even released yet (the Jaguar Type-00 for example), with data-driven results and confidence ranges. To our knowledge, no one matches the breadth and accuracy of Brego here. The same goes for luxury vehicles, new vehicles and specialist asset types such as static holiday homes and HGVs; because of the nature of neural networks, we can value these assets with the same level of precision, and we provide explainability so lenders and dealers can trust the output.

“I do see a landscape where more companies adopt AI and learn how to utilise it beyond the Large Language Model (LLM) integrations we’re seeing at the moment. When the industry gets to that point, data will be the primary differentiator. Companies that have more, clean data to train models will pull ahead. We’re already seeing leading AI companies purchase training-data businesses for millions of pounds and dollars, and that dynamic will extend to industries where AI is applicable.”

Looking ahead to 2030 and beyond, you’re actively modelling dramatic market shifts including government mandates and changing fuel preferences. What role do you see data quality playing as the primary differentiator when more companies adopt AI, and how are you positioning the company to maintain its competitive edge?